Cure rates in childhood acute lymphoblastic leukemia (ALL) have increased to over 80%. Current therapies depend heavily upon the intense use of glucocorticoids, particularly dexamethasone, in addition to multiple other antileukemic agents. The major dose-limiting adverse effect of glucocorticoids in modern ALL clinical trials is glucocorticoid-induced osteonecrosis or avascular necrosis. Glucocorticoid-induced osteonecrosis has long been known to be related to glucocorticoid exposure in adults and children treated for solid organ transplant, nephropathies, asthma, arthritis, and other autoimmune diseases. Our group and others have identified putative treatment-related and host-related risk factors influencing glucocorticoid-induced osteonecrosis. However, any changes to therapy to attempt to decrease the risk of this adverse effect must be weighed against the possible attenuation of desired antileukemic effects. Thus, preclinical models that allow for the characterization of treatment-related and host-related risk factors for osteonecrosis are needed, but they must be coupled with studies on how those same factors affect effectiveness (i.e., relapse risk). Our group has spent the last two years developing the first murine model for glucocorticoid-induced osteonecrosis, as the necessary platform for extending our prior studies of risk factors in children with ALL. Using our clinical data as the impetus for our laboratory studies in murine models, our goal is to fully elucidate treatment-related and host-related risk factors for glucocorticoid-induced osteonecrosis, and to evaluate the influence of these risk factors on antileukemic effectiveness. These translational research questions cannot be addressed in patients. Three aims are proposed: to compare the frequency of osteonecrosis following discontinuous vs continuous dexamethasone in a multiagent regimen; to compare the antileukemic effects of discontinuous vs continuous dexamethasone in two murine models of ALL; and to compare the osteonecrotic vs antileukemic effects of dexamethasone in mice that are wild-type, hemizygous, and homozygous deficient for germline genetic defects identified as related to clinical susceptibility to osteonecrosis. Our long term objective is to design less toxic glucocorticoid-containing regimens for ALL that do not compromise desired antileukemic effectiveness.